WO2023221182A1 - Pet feed recommendation method and system based on artificial intelligence - Google Patents

Abstract

A pet feed recommendation method and system based on artificial intelligence, relating to the technical field of artificial intelligence. The method comprises: acquiring pet-related information by using a pet data acquisition device to obtain basic pet information; analyzing feeding requirements respectively from a plurality of dimensions of age information, breed information, weight information, and special body condition information according to the basic pet information, and determining pet feeding requirement information; and performing traversal analysis in a feed ingredient list library on the basis of the pet feeding requirement information, and determining recommended feed information and providing feedback.

Description

An artificial intelligence-based pet food recommendation method and system Technical field

The present invention relates to the field of artificial intelligence technology, and specifically relates to a method and system for recommending pet food based on artificial intelligence.

Background technique

As the material life of our country's citizens gradually becomes more fulfilling, more and more people are beginning to pursue spiritual satisfaction, and pets have made a huge contribution in meeting people's spiritual needs. Take the most classic canine and feline pets as examples. Because of their furry touch and strong interactivity, they can psychologically promote the secretion of progesterone and make pet owners feel happy. Therefore, more and more people are becoming pet owners. The ranks of pet owners began to raise pets. With the continuous expansion of the pet market, reptiles and amphibians gradually appeared in the list of pet owners. At the same time, not all pet owners have the same feeding capabilities and feed selection and matching capabilities as pet doctors or breeders.

Since there are many types of pet feed, and the feeding and nutritional requirements of each type of pet are different, in the existing technology, pet owners cannot select suitable feed for feeding according to the survival and nutritional needs of their pets, thereby affecting the health of their pets and even shortening their lifespan. technical issues.

Contents of the invention

This application provides a method and system for recommending pet feed based on artificial intelligence, which is used to solve the problem in the existing technology that pet owners cannot select suitable feed for feeding according to the survival and nutritional needs of their pets, thereby affecting the health status of their pets or even Technical issues that shorten the lifespan of pets.

In view of the above problems, this application provides an artificial intelligence-based pet food recommendation method and system.

The first aspect of this application provides a pet feed recommendation method based on artificial intelligence. The method includes: using a pet data collection device to collect pet-related information to obtain basic pet information; Carry out feeding demand analysis from several dimensions such as information, breed information, weight information, and special physique information to determine pet feeding demand information; conduct traversal analysis in the feed ingredient list library based on pet feeding demand information, determine recommended feed information, and recommend the feed information Give feedback.

The second aspect of this application provides a pet feed recommendation system based on artificial intelligence. The system includes: a pet data collection module, used to collect pet-related information using pet data collection equipment to obtain basic pet information, Wherein, the basic pet information includes age information, breed information, weight information, and special constitution information; the feeding needs analysis module is used to select from age information, breed information, weight information, and special constitution information according to the basic pet information. Dimensionally carry out feeding demand analysis to determine pet feeding demand information; the feed recommendation feedback module is used to perform traversal analysis in the feed ingredient list library based on the pet feeding demand information, determine recommended feed information, and feedback the recommended feed information .

One or more technical solutions provided in this application have at least the following technical effects or advantages:

The method provided by the embodiment of this application collects pet-related information by using pet data collection equipment to obtain pet basic information; based on the pet's basic information, the breeding needs are analyzed from several dimensions: age information, breed information, weight information, and special physical information. , after the pet feeding demand information is determined, a traversal analysis is performed in the feed ingredient list library based on the pet feeding demand information, the recommended feed information is determined and feedback is provided. This application collects relevant basic information about pets to provide a reference information basis for subsequent analysis and acquisition of pet feeding needs. Comprehensive analysis of the current nutritional needs and growth needs of pets based on the pet's age information, breed information, weight information, and special physical information to avoid one-sided analysis of pet feed needs from a single perspective, resulting in nutritional deficiencies in the recommended feed. Based on big data, pet feeding needs are matched to obtain corresponding pet feed products whose feed ingredients meet the analyzed pet feed needs. The actual recommended pet feed is consistent with pet needs, meets the growth needs and nutritional needs of pets, and provides pets with Technical effects of health protection.

Description of the drawings

Figure 1 is a schematic flow chart of an artificial intelligence-based pet food recommendation method provided by this application;

Figure 2 is a schematic flow chart of constructing a feed ingredient list library in an artificial intelligence-based pet feed recommendation method provided by this application;

Figure 3 is a schematic flow chart of obtaining matching and adjusted feed information in an artificial intelligence-based pet food recommendation method provided by this application;

Figure 4 is a schematic structural diagram of an artificial intelligence-based pet food recommendation system provided by this application;

Explanation of reference signs: pet data collection module 11, feeding needs analysis module 12, feed recommendation feedback module 13

Detailed ways

This application provides a method and system for recommending pet feed based on artificial intelligence, which is used to solve the problem in the existing technology that pet owners cannot select suitable feed for feeding according to the survival and nutritional needs of their pets, thereby affecting the health status of their pets or even Technical issues that shorten the lifespan of pets.

In response to the above technical problems, the general idea of the technical solution provided by this application is as follows:

Collect basic pet information that is relevant to the pet's growth status and health and nutritional needs, and analyze pet feeding needs based on the basic pet information, obtain the current pet nutritional needs and growth needs, and match pet feeding needs based on big data. Obtain corresponding pet feed products whose feed ingredients meet the needs of the pet feed obtained through analysis. It avoids a one-sided analysis of pet needs, which may lead to nutritional deficiencies in the recommended feed. It achieves the technical effect of actually recommending pet feed that matches the needs of pets and meets the growth and nutritional needs of pets.

Embodiment 1

As shown in Figure 1, this application provides an artificial intelligence-based pet food recommendation method, which includes:

S100: Use pet data collection equipment to collect pet-related information and obtain basic pet information, where the basic pet information includes age information, breed information, weight information, and special physical information;

Specifically, the pets are creatures kept by humans to satisfy spiritual needs, covering a variety of biological types such as fish, reptiles, amphibians, and insects. The pet data collection device is an intelligent data collection device with text recognition function, image collection function and gravity sensing function. The special physical condition information refers to the specific physical condition of a specific pet compared with other pets of the same category, or the general special physical condition of pets of the same breed, such as Garfield cats with clogged tear ducts and Border Collies with fragile gastrointestinal tracts.

There are three methods for the pet data collection device to obtain basic pet information to be recommended as feed:

First, pet owners input the historical medical records of their pets into the pet data collection device, and the pet data collection device identifies and calculates the pet's current age information, special physical information, and quality information from the pet's historical medical records;

Second, the pet owner directly inputs the pet's basic information including age, special physique, and breed into the pet data collection device;

Third, the pet data collection device obtains the image information of the pet to be recommended as feed, performs image feature recognition, obtains the pet's coat color, facial features, ear shape and other information, and determines that the pet includes age information, special physical information and Basic pet information for breed information.

Use the pet data collection device to collect pet-related information to obtain the pet's basic information including age information, breed information, weight information, and special physique information.

S200: According to the basic information of the pet, conduct a breeding demand analysis from several dimensions: age information, breed information, weight information, and special physical information, and determine the pet feeding demand information;

Specifically, it should be understood that pets of different ages have different growth nutritional requirements and feeding amount references, so different pet feeds need to be selected according to age. Different breeds of pet dogs have different endocrine characteristics, hair follicle structures, and diseases, so they need to choose feeds with different nutritional emphasis for feeding. Different pet dogs may have different special physiques due to genetic and hereditary reasons. Therefore, in order to avoid growth and development disorders of pet dogs caused by genetic diseases or special physiques caused by genes, it is necessary to choose feeds that contain ingredients that can relieve special physiques and be fed.

Take the raising of pet dogs as an example. The puppy stage is an important part of the growth and development of pet dogs. It requires high protein and energy content in the feed. At the same time, due to the weak gastrointestinal function of puppies, the feed is required to be easy to digest and absorb. In the adult stage, pet dogs When the dog is mature, nutritionally balanced feed should be used for feeding at this time. At the same time, in order to protect the teeth of adult dogs, the feed is required to be hard to grind teeth. At the same time, fish and sheep are regularly supplemented for nutritional enhancement; in the old age, pet dogs have weakened endocrine ability and calcium deficiency. The intake of calcium gradually decreases and the loss increases. At this time, it is necessary to choose feeds with higher calcium content for feeding. At the same time, in order to avoid constipation caused by the pet's low activity and weakened digestion ability, it is necessary to choose feeds containing an appropriate amount of plant fiber.

Compared with other dog breeds, Poodles are more likely to be troubled by dental calculus, and due to their special composite hair follicle structure, their hair is prone to mattness. Therefore, for Poodles, you need to focus on choosing pet food with a beautiful hair and tooth-protecting formula. Schnauzers Compared with other dog breeds, the urethra is sensitive and prone to skin diseases. Therefore, Schnauzers need to focus on choosing pet food formulated to maintain urinary tract and protect fur.

Garfield cats have a special physique that is prone to lacrimal gland inflammation, and need to be fed feeds with ingredients that inhibit lacrimal gland inflammation or have health-care effects; Corgis have a special physique that is prone to calcium deficiency and do not like to drink water, so they need to be fed canned foods with high calcium and water content. feed to avoid corgi constipation and weak bones.

Conduct a comprehensive feeding demand analysis from the age information, breed information, weight information, and special physical information of the pet to be recommended for feed, and determine the feeding demand information of the pet, for example, for puppies who are in the puppy stage and suffer from hereditary intestinal diseases. Feeding needs of sick corgis.

S300: Perform traversal analysis in the feed ingredient list library based on the pet feeding demand information, determine recommended feed information, and feed back the recommended feed information.

Specifically, it should be understood that different pet feeds often have different formulas to meet the needs of raising different types of pets. Based on different ingredient formulas, different types of pet feeds have different protein, carbohydrates and other nutritional components.

The feed ingredient list database is a large-scale database constructed based on big data collection and covers the vast majority of pet feeds currently circulating in the market. Based on the pet feeding demand information, a traversal analysis is performed in the feed ingredient list database to determine that the pet feed requirements are met. Feed single or multiple feed products corresponding to the demand information, and provide the recommended feed product information to pet owners for pet feed recommendation feedback, so as to provide pet owners with pet feed that facilitates the healthy growth of pets and meets their nutritional and health needs.

This application collects relevant basic information about pets to provide a reference information basis for subsequent analysis and acquisition of pet feeding needs. Comprehensive analysis of the current nutritional needs and growth needs of pets based on the pet's age information, breed information, weight information, and special physical information to avoid one-sided analysis of pet feed needs from a single perspective, resulting in nutritional deficiencies in the recommended feed. Based on big data, pet feeding needs are matched to obtain corresponding pet feed products whose feed ingredients meet the analyzed pet feed needs. The actual recommended pet feed is consistent with pet needs, meets the growth needs and nutritional needs of pets, and provides pets with Technical effects of health protection.

Further, as shown in Figure 2, the method step S300 provided by this application also includes:

S310: Analyze each feed ingredient through a near-infrared spectrum analyzer to obtain feed monitoring ingredient information;

S320: Determine the feed composition information through the feed introduction information;

S330: Perform correlation analysis on the feed monitoring component information and the feed component information, map the corresponding relationship between the feed monitoring component information and the feed component information, and construct the feed component list library.

Specifically, the feed composition information is the nutritional composition table and ingredient list corresponding to the pet feed produced, provided by the feed manufacturer. The feed monitoring ingredient information is the crude fiber vegetable, poultry liver, and meat product ingredient information that specifically corresponds to the fat, protein, plant fiber, energy and other information in the feed nutritional ingredient table. In an ideal state, the feed monitoring component information obtained through experiments is consistent with the feed composition information provided by the feed manufacturer.

The principle of the near-infrared spectrum analyzer to analyze feed ingredients is to use near-infrared diffuse reflectance spectroscopy technology to scan and acquire the near-infrared diffuse reflectance spectrum of the feed. The specific method is to take an appropriate amount of pet feed, remove the water in a dryer, grind it through ultra-fine grinding or set the grinding coefficient of other particle size levels to crush the feed, and set the mesh standard to sieve the crushed feed. Put an appropriate amount of feed into the rotating sample cup, and scan it on the machine (each sample is loaded and scanned repeatedly for 1 to 16 times. This embodiment does not limit the number of infrared scans. It can be set according to actual needs during the specific implementation process. ), and then calculate the average spectrum to obtain the near-infrared diffuse reflectance spectrum of the crushed feed sample.

Based on the obtained near-infrared diffuse reflectance spectrum of the feed, feed component feature analysis is performed to determine the feed monitoring component information. Perform correlation analysis on the feed monitoring component information and the feed component information, and map the corresponding relationship between the feed monitoring component information and the feed component information to construct the feed component list library.

This embodiment obtains the monitored ingredient information of pet feed, compares and corrects the feed ingredient data provided by the merchant, and inputs the constructed feed ingredient list library to avoid making feed recommendations to pet owners by solely referring to the feed information provided by the merchant, resulting in recommendation results. The problem of inaccuracy has been achieved, and the technical effect of improving the matching degree between the recommended pet feed and the pet feed demand has been achieved.

Further, each feed component is analyzed by a near-infrared spectrum analyzer to obtain feed monitoring component information. The method step S310 provided by this application also includes:

S311: Collect images of feed to obtain feed images;

S312: Reconstruct the feed image through Fourier transform to obtain a Fourier transform spectrogram;

S313: Perform crude fat and crude fiber feature analysis based on the Fourier transform spectrum chart to determine the feed monitoring component information.

Specifically, obtain a specific pet feed to be monitored, take an appropriate amount of pet feed, and grind it through ultra-fine grinding or set the grinding coefficient of other particle size levels based on the dryer to remove the moisture, and carry out the mesh screening standard Set up to carry out sieving processing of crushed feed. Put an appropriate amount of feed into the rotating sample cup, scan it on the machine, and complete image collection to obtain the feed image.

In this embodiment, the Fourier transform is used to characterize the molecular structure and material chemical composition information in the feed sample. The Fourier transform near-infrared spectroscopy method can be used to determine the characteristics of the content of biological macromolecules such as protein and starch. The near-infrared diffuse reflection spectrum image of the feed sample is reconstructed through Fourier transformation to obtain a Fourier transform spectrum diagram; according to the Fourier transform spectrum diagram, the fixed frequency and amplitude represented by the power on the spectrum diagram are obtained. The sine wave image of value and phase is filtered based on the sine wave image, and then the Fourier transform spectrogram is inversely transformed to obtain an identifiable image, thereby improving the identification of components such as crude fat, crude fiber, and protein. The resolution of the analysis determines the feed monitoring ingredient information.

This embodiment is based on Fourier transform near-infrared spectroscopy to quickly obtain the ingredients of the pet feed to be tested, improves the efficiency of pet feed component analysis, and provides a data basis for subsequent comparisons with pet feed formula components provided by manufacturers. , indirectly achieving the technical effect of improving the accuracy of pet food recommendation.

Further, according to the pet's basic information, the feeding needs are analyzed from several dimensions including age information, breed information, weight information, and special physical information, and the pet feeding demand information is determined. The method step S200 provided by this application includes:

S210: Analyze feeding needs from the age information, breed information, weight information, and special constitution information respectively, and obtain age feeding needs, breed feeding needs, weight feeding needs, and physical feeding needs;

S220: Calculate the impact weight of the age information, breed information, weight information, and special physique information on feeding needs;

S230: Perform weighting processing on the age feeding demand, breed feeding demand, weight feeding demand, and physical feeding demand based on the influence weight to obtain the pet feeding demand information.

Specifically, the age feeding requirements are based on the characteristics that pets of different ages have different growth nutritional requirements and feeding amount references. Different pet feeds are selected according to different ages. The breed feeding requirements are based on the fact that pet dogs of different breeds have different endocrine characteristics, hair follicle structures, and disease characteristics. According to different breeds, feeds with different nutritional emphasis should be selected for feeding. The weight feeding requirement is the current feed amount and frequency for the pet determined based on the specific pet's eating habits, feeding frequency and the average weight status of pets of the same type at the current age. The special constitution feeding requirements are based on the characteristics of different special constitutions based on the genes and genetic disease factors of different pet dogs. In order to avoid the growth and development disorders of pet dogs caused by genetic diseases or special constitutions caused by genes, select products that can alleviate the special constitution. Feed with health-care ingredients.

The feeding needs are analyzed from the dimensions of age information, breed information, weight information, and special constitution information respectively, and the corresponding age feeding needs, breed feeding needs, weight feeding needs, and physical feeding needs are obtained.

It should be understood that different pets are affected by age information, breed information, weight information, and special physical information due to their different age information, breed information, weight information, and special physical information, and have different degrees of interference with the normal growth and health of pets. , so calculations are performed to determine the impact weight of the age information, breed information, weight information, and special physique information on specific pet raising needs.

Specifically, the weight value for weight distribution of different pet basic information can directly use the weight value provided by the existing technology as the weight distribution result, or can also be obtained by using methods such as analytic hierarchy process, fuzzy method, fuzzy analytic hierarchy process, and expert evaluation method. The weight values corresponding to different pet basic information are obtained as the weight distribution results. The age feeding demand, breed feeding demand, weight feeding demand, and constitution feeding demand are weighted based on the influence weight to obtain the pet feeding demand information.

According to the characteristics that different pet basic information has different effects on the health status and growth status of specific pets, the embodiments of this application weight the age feeding needs, breed feeding needs, weight feeding needs, and physical feeding needs based on the influence weights. Through processing, pet feeding demand information that is more representative of pet feeding needs is obtained, and the technical effect of obtaining reference feeding information that can accurately reflect pet nutritional needs is achieved.

Further, the method step S200 provided by this application also includes:

S240: Construct an evaluation data set based on the age information, breed information, weight information, and special physique information;

S250: Obtain preset N evaluation channels, where N is a positive integer greater than 5, randomly group the constructed evaluation data sets, and determine N groups of evaluation data groups;

S260: Input the N groups of evaluation data groups into the preset N evaluation channels respectively, and obtain the weight evaluation results of each parameter of the age information, breed information, weight information, and special physical information through feedback from the evaluation channels;

S270: Calculate the average value according to the weight evaluation results of each parameter, and determine the influence weight.

Specifically, in this embodiment, the evaluation channel is a data channel with an information isolation function. There is no mutual interference or data pollution between data in each channel. Multiple groups of pet basic data weights are implemented based on the evaluation channel. Assign groups of information. In order to obtain a more accurate distribution of influence weights, the basic pet information weight evaluations of experts in the pet field and professional pet raising platforms can be adopted through public channels, or the basic pet information weight evaluations given by some non-specific pet raising groups can be captured based on big data. , thereby obtaining multiple sets of pet basic information weight data, forming an evaluation data set containing multiple sets of evaluation data. Taking pet basic information as the unit, a set of pet basic information weight distribution evaluation data corresponds to a weight distribution channel, and N evaluation channels are preset. The constructed evaluation data sets are randomly grouped to determine N groups of evaluation data groups; The N groups of evaluation data groups are input to the preset N evaluation channels, and the weight evaluation results of each parameter of the age information, breed information, weight information, and special physical information are obtained through feedback from the evaluation channels; according to the weight evaluation results of each parameter Calculate the average and determine the impact weight.

This embodiment collects multiple groups of pet basic information weight distribution evaluation data groups to isolate information, so that the obtained pet basic information weight distribution relationship is less subjective and can better reflect the importance of each basic information in pet nutrition and growth needs.

Further, the method steps provided by this application also include:

S510: Obtain pet growth and feeding habits information;

S520: Evaluate the pet's growth and feeding habit information based on the pet's basic information to obtain the feeding effect evaluation results;

S530: When the feeding effect evaluation result does not meet the feeding standard, generate feeding adjustment information based on the pet growth and feeding habit information and the recommended feed information.

Specifically, the information on the pet's growth and feeding habits includes the pet owner's daily average feed amount and the proportion of feed for each meal during the pet's growth process, and also includes the pet owner's "nutrition" of the pet. information such as the frequency of addition of additional nutritional supplements such as "cream" and "vitamin tablets", as well as the amount and frequency of pet exercise. Evaluate the pet's growth and feeding habit information based on the pet's basic information, determine whether the pet owner's feeding of the pet poses a risk of causing nutritional deficiencies or excessive obesity in the pet, and obtain the feeding effect evaluation results.

When the feeding effect evaluation result does not meet the feeding standards, feeding adjustment information is generated based on the pet growth and feeding habit information and the recommended feed information, and the pet owner refers to the feeding adjustment information to improve the feeding habits of the pet and avoid Health risks to pets caused by overfeeding or underfeeding.

This embodiment combines the feed usage frequency of specific pet owners during daily feeding and the feeding status of other nutritional enhancement substances to make recommendations for adjusting the feeding habits of the generated recommended feeds, thereby further improving the growth status of pets fed with recommended feeds. Healthy and good technical results.

Further, as shown in Figure 3, the method step S520 provided by this application also includes:

S521: When the feeding effect evaluation result meets the feeding standard, determine whether the recommended feed information satisfies the pet growth and feeding habit information;

S522: When not satisfied, obtain difference information based on the recommended feed information and the pet growth and feeding habit information;

S523: Determine whether the parameters corresponding to the difference information are included in the parameters of the feed ingredient list library. When included, perform a traversal comparison from the feed ingredient list library based on the difference information and the recommended feed information. , obtain matching adjustment feed information.

Specifically, when the feeding effect evaluation results meet the feeding standards, that is, the pet owner's daily feeding habits have no risk of affecting the pet's health and normal growth, it can be further judged that the recommended feed information is suitable for the current pet owner's feeding habits. Whether daily feeding can maintain the pet's health or further optimize the pet's health. That is, the current recommended feed satisfies the pet's growth and feeding habits. When the pet's growth and feeding habits are not satisfied, the difference information is obtained based on the recommended feed information and the pet's growth and feeding habits information; it is determined whether the difference information corresponds to Whether the parameters are included in the parameters of the feed ingredient list library, when included, a traversal comparison is performed from the feed ingredient list library based on the difference information and the recommended feed information to obtain matching and adjusted feed information.

This embodiment simulates and compares the pet owner's daily feeding habits with the current recommended feed obtained based on the pet's basic information to determine whether it is necessary to supplement the original recommended feed to match the pet's original eating habits. It has achieved the technical effect of further improving the fit between the recommended feed and the pet’s eating habits.

Embodiment 2

Based on the same inventive concept as the artificial intelligence-based pet food recommendation method in the previous embodiment, as shown in Figure 4, this application provides an artificial intelligence-based pet food recommendation system, wherein the system includes:

The pet data collection module 11 is used to collect pet-related information using pet data collection equipment and obtain basic pet information, where the basic pet information includes age information, breed information, weight information, and special physique information;

The feeding needs analysis module 12 is used to analyze the feeding needs from several dimensions including age information, breed information, weight information, and special physical information according to the basic pet information, and determine the pet feeding need information;

The feed recommendation feedback module 13 is used to perform traversal analysis in the feed ingredient list library based on the pet feeding demand information, determine recommended feed information, and feedback the recommended feed information.

Further, the recommended feed feedback module also includes:

The feed ingredient analysis unit is used to analyze each feed ingredient through a near-infrared spectrum analyzer to obtain feed monitoring ingredient information;

The feed component identification unit is used to determine the feed component information through the feed introduction information;

A recommended feed list construction unit is used to perform correlation analysis on the feed monitoring component information and the feed component information, and map the corresponding relationship between the feed monitoring component information and the feed component information to construct the The feed ingredient list library.

Further, the feed ingredient analysis unit also includes:

A feed image acquisition unit is used to collect images of feed to obtain feed images;

A feed image reconstruction unit, configured to reconstruct the feed image through Fourier transform to obtain a Fourier transform spectrogram;

An ingredient feature analysis unit is used to perform crude fat and crude fiber feature analysis based on the Fourier transform spectrum chart to determine the feed monitoring ingredient information.

Further, the feeding needs analysis module also includes:

The situation analysis unit is used to analyze the feeding needs from the age information, breed information, weight information, and special constitution information respectively, and obtain the age feeding needs, breed feeding needs, weight feeding needs, and physical feeding needs;

A weight allocation unit is used to calculate the weight of the influence of the age information, breed information, weight information, and special physical information on the feeding needs;

A weighting processing unit is configured to perform weighting processing on the age feeding demand, breed feeding demand, weight feeding demand, and physical feeding demand based on the influence weight, to obtain the pet feeding demand information.

Further, the weight allocation unit also includes:

An evaluation data acquisition unit is used to construct an evaluation data set based on the age information, breed information, weight information, and special physical information;

An evaluation channel construction unit is used to obtain preset N evaluation channels, where N is a positive integer greater than 5, randomly group the constructed evaluation data sets, and determine N groups of evaluation data groups;

An evaluation result obtaining unit is used to input the N groups of evaluation data groups into the preset N evaluation channels respectively, and obtain the weights of each parameter of the age information, breed information, weight information, and special physique information through feedback from the evaluation channels. Evaluation results;

The influence weight obtaining unit is used to calculate the average value according to the weight evaluation results of each parameter and determine the influence weight.

Further, the feed recommendation feedback module also includes:

The feeding habit acquisition unit is used to obtain pet growth and feeding habit information;

A feeding habit evaluation unit, used to evaluate the pet's growth and feeding habit information based on the pet's basic information to obtain a feeding effect evaluation result;

An evaluation result processing unit is configured to generate feeding adjustment information based on the pet growth and feeding habit information and the recommended feed information when the feeding effect evaluation result does not meet the feeding standard.

Further, the evaluation result processing unit also includes:

A feeding difference generation unit is used to obtain difference information based on the recommended feed information and the pet growth and feeding habit information when the conditions are not met;

The feed adjustment processing unit is used to determine whether the parameters corresponding to the difference information are included in the parameters of the feed ingredient list library. When included, based on the difference information and the recommended feed information, the parameters corresponding to the difference information are included in the parameters of the feed ingredient list library. Perform traversal comparison to obtain matching and adjusted feed information.

In summary, it can be seen that the artificial intelligence-based pet feed recommendation method and system of the present invention is used to provide a reference information basis for subsequent analysis and acquisition of pet feeding needs based on the relevant basic information of the pet. Based on the pet's age information, breed information, Comprehensive analysis of current pet nutritional needs and growth needs using weight information and special physical information to avoid one-sided analysis of pet feed needs from a single perspective, resulting in nutritional deficiencies in the recommended feed. Based on big data, pet feeding needs are matched to obtain corresponding pet feed products whose feed ingredients meet the analyzed pet feed needs. The actual recommended pet feed is consistent with pet needs, meets the growth needs and nutritional needs of pets, and provides pets with Technical effects of health protection.

Based on the above-mentioned specific embodiments of the present invention, any improvements and modifications made by those skilled in the art to the present invention without departing from the principles of the present invention shall fall within the scope of patent protection of the present invention.

Claims (8)

1. A method for recommending pet food based on artificial intelligence, characterized in that the method includes:

   Use pet data collection equipment to collect pet-related information to obtain basic pet information, where the basic pet information includes age information, breed information, weight information, and special physical information;

   According to the basic pet information, the breeding needs are analyzed from several dimensions: age information, breed information, weight information, and special physical information, and the pet feeding needs information is determined;

   Based on the pet feeding demand information, a traversal analysis is performed in the feed ingredient list library to determine recommended feed information, and the recommended feed information is fed back.
2. The method of claim 1, wherein the method includes:

   Analyze each feed ingredient with a near-infrared spectrometer to obtain feed monitoring ingredient information;

   Determine feed composition information through feed introduction information;

   Perform correlation analysis on the feed monitoring component information and the feed component information, and map the corresponding relationship between the feed monitoring component information and the feed component information to construct the feed component list library.
3. The method of claim 2, wherein each feed ingredient is analyzed by a near-infrared spectrometer to obtain feed monitoring ingredient information, including:

   Collect images of feed to obtain feed images;

   Perform image reconstruction through Fourier transform on the feed image to obtain a Fourier transform spectrogram;

   Carry out crude fat and crude fiber characteristic analysis based on the Fourier transform spectrum chart to determine the feed monitoring component information.
4. The method of claim 1, wherein the pet's basic information is analyzed from several dimensions including age information, breed information, weight information, and special physique information to determine the pet's feeding needs information, including :

   The feeding needs are analyzed from the dimensions of age information, breed information, weight information, and special constitution information respectively, and the age feeding needs, breed feeding needs, weight feeding needs, and physical feeding needs are obtained;

   Calculate the impact weight of the age information, breed information, weight information, and special physique information on feeding needs;

   The age feeding demand, breed feeding demand, weight feeding demand, and constitution feeding demand are weighted based on the influence weight to obtain the pet feeding demand information.
5. The method of claim 4, wherein the method includes:

   Construct an evaluation data set based on the age information, breed information, weight information, and special physical information;

   Obtain preset N evaluation channels, where N is a positive integer greater than 5, randomly group the constructed evaluation data sets, and determine N groups of evaluation data groups;

   The N groups of evaluation data groups are respectively input into the preset N evaluation channels, and the weight evaluation results of each parameter of the age information, breed information, weight information, and special physical information are obtained through feedback from the evaluation channels;

   Calculate the average value based on the weight evaluation results of each parameter to determine the influence weight.
6. The method of claim 1, wherein the method includes:

   Obtain information on pet growth and feeding habits;

   Evaluate the pet's growth and feeding habit information based on the pet's basic information to obtain the feeding effect evaluation results;

   When the feeding effect evaluation result does not meet the feeding standard, feeding adjustment information is generated based on the pet growth and feeding habit information and the recommended feed information.
7. The method of claim 6, wherein when the feeding effect evaluation result meets the feeding standard, it is determined whether the recommended feed information satisfies the pet growth and feeding habit information;

   When not satisfied, obtain difference information based on the recommended feed information and the pet growth and feeding habit information;

   Determine whether the parameters corresponding to the difference information are included in the parameters of the feed ingredient list library. When included, traverse and compare from the feed ingredient list library based on the difference information and the recommended feed information to obtain Match adjusted feed information.
8. A pet food recommendation system based on artificial intelligence, characterized in that the system includes:

   The pet data collection module is used to collect pet-related information using pet data collection equipment and obtain basic pet information, where the basic pet information includes age information, breed information, weight information, and special physique information;

   The feeding needs analysis module is used to analyze the feeding needs from several dimensions including age information, breed information, weight information, and special physical information based on the basic information of the pet, and determine the pet feeding need information;

   The feed recommendation feedback module is used to perform traversal analysis in the feed ingredient list library based on the pet feeding demand information, determine recommended feed information, and feedback the recommended feed information.

Images